Elements appear too close to the edge?

• May 9, 2013 - 10:28

1. Open attached score (produced in 1.3).

Discussion: I don't know the rules, but if you look at the first notes in a bar that isn't the beginning of a line in 2.0, the elements seem too close to the edge. What do others think?

I include several PNGs, because stem direction is different in 1.3 and 2.0. Also see LilyPond's result.

Using MuseScore 2.0 Nightly Build (bc04561) - Mac 10.7.5.


I didn't know, but here's what I've been able to find out.

According to Alfred's Dictionary of Music Notation it depends on the density of the notes in a measure. If you have a measure of 32nd notes then use less space after the barline. If you have a measure quarter notes then use slightly more space.

Finale uses the same space regardless of density (slightly more than the width of a notehead).

MPA speaks strongly against adding any extra spacing to the barline ("since it has no rhythmic value") and prefers that it be spaced between "close to the first note of the measure after that note has been time placed according to the preceding note".

Elaine Gould doesn't seem to speak to the matter as far as I can tell.

In reply to by David Bolton

I tried to measure the space between barline and first note in both Sib and Fin and they seem to use 1.5space.

One interesting thing, Sibelius uses this space to put the accidental if any. The accidental ends up closer to the barline.
While Finale moves the note a space on the right to put the accidental. The accidental ends up at 1.25 space from the barline.

Sibelius on the left, Finale on the right.

In MuseScore, the default is currently to put the notehead at 1.2space. When an accidental is added, the notehead is shifted, the accidental ends up at 1.2 space too.

Lilypond anyone?

Attachment Size
08.png 151.61 KB

In reply to by [DELETED] 5

I do prefer the space between the barline and note to be partially taken up by the accidental (as in the Sibelius example). In examining older scores, this seems to be common practice. To me, it is a more elegant and efficient use of space. The same for appogiaturas at the beginnings of measures.

See bars 2 and 4 of the attached (2.0, Gonville).

Attachment Size
Mesta o dio-01.png 133.03 KB

In reply to by chen lung

The change I proposed is simply to fix 2.0 to render consistently. Right now, 1.3 renders 1.3 scores with 1.20sp between the barline and first note by default. And 2.0 renders 2.0 scores with 1.20sp between the barline and first note by default. However, 2.0 renders 1.3 scores with only 0.60sp by default. My change fixes this - 2.0 will now render 1.3 scores with 0.60sp by default.

Note that you could override this default in 1.3, and if you did so, 2.0 will honor your customization. It was only 1.3 scores that did *not* override the default that 2.0 disoplays incorrectly, so it is only these scores my change affects.

This change in no way addresses the question of how to deal with accidentals. That behavior is unaffected by my fix.

In reply to by chen lung

Regarding the rest of this discussion:

The but about how much to increase amend the configured barline to note distance on the fly for measures starting with an accidental is a separate issue. And actually, it looks like that is *already* implemented. It is in 1.3 at least as well as the current development builds. Try it for yourself: put a quarter note at the beginning of a measure, observe the amount of space between it and the left barline, then raise the note a half step. You'll see the note moves only a little to the right; the accidental mostly takes up the space that was there already. In other words, MuseScore looks like the Sibelius example above. Not sure what lasconic was looking at when he said MuseScore put the accidental at 1.2sp - it definitely does not that I can see. Maybe that was temporarily broken?

The idea of varying the barline spacing according to density seems too vague to be useful. And the MPA guideline saying no extra space should be allocated is nonsensical to me. Maybe it made more sense in context? But it sounds to me hat they are suggesting you space the music as if there no barline, then just stick the barline in there somewhere, and that's definitely *not* how any music I know is typeset. In fact would look ridiculous for note values shorter than half notes, so I'm thinking this is being misinterpreted here (by either David or me).

Gould *does* address this, quite specifically, on pages 42-43. She calls 1sp on either side of the barline. Regarding accidentals, she says they "may" be closed up to within 0.5sp "where space is limited". But she clearly regards this as optional, as most of her examples show this *not* happening, and the one example where she does this she labels as "acceptable in cramped conditions". But there is 1sp distance from barline to accidental pretty consistently in her examples throughout the book.

I wouldn't be opposed to changing the 2.0 default from 1.2sp to 1sp in accordance with Gould, assuming this resulted in decent spacing in the presence of accidentals (further tweaks might be required).

Do you still have an unanswered question? Please log in first to post your question.